. 2012 Dec 4;24(2):567–580. doi: 10.1007/s00198-012-2224-2

Table 4.

Hazard ratios and 95 % confidence intervals for calcium plus vitamin D supplementation from the WHI CaD trial and Observational Study according to years from supplement initiation: invasive cancer

Years since CaD initiation	CaD trial				Observational study		Combined trial and OS
	All participants		No personal supplements^a		Observational study		All participants		No personal supplements^a
	HR	95 % CI	HR	95 % CI	HR	95 % CI	HR	95 % CI	HR	95 % CI
	Colorectal cancer
<2	0.89	0.57,1.38	0.71	0.35,1.44	0.94	0.23,3.87	0.92	0.60,1.40	0.75	0.39,1.45
2–5	1.00	0.71,1.41	0.75	0.45,1.24	0.80	0.39,1.65	1.02	0.74,1.41	0.78	0.50,1.21
>5	1.30	0.88,1.92	0.99	0.56,1.77	0.83	0.60,1.14	1.23	0.87,1.74	0.90	0.56,1.45
Trend test^b	0.19		0.44		0.96		0.26		0.57
HR in OS/HR in trial^d							0.69	0.45,1.07	0.94	0.55,1.59
Overall HR^d	1.06	0.85, 1.32	0.81	0.58, 1.13	0.83	0.61, 1.12
	Breast cancer
<2	1.00	0.79,1.27	0.98	0.68,1.42	0.90	0.44,1.83	0.97	0.78,1.22	0.88	0.64,1.22
2–5	0.98	0.82,1.18	0.75	0.56,1.00	1.05	0.78,1.41	0.95	0.81,1.12	0.75	0.59,0.95
>5	0.89	0.72,1.11	0.73	0.52,1.02	1.14	1.00,1.30	0.95	0.80,1.14	0.80	0.62,1.02
Trend test^b	0.45		0.26		0.42		0.89		0.87
HR in OS/HR in trial^c							1.18	0.96,1.45	1.42	1.09,1.84
Overall HR^d	0.96	0.85, 1.08	0.80	0.66, 0.96	1.12	0.99, 1.28
	Total invasive cancer
<2	0.96	0.83,1.12	0.96	0.76,1.22	0.87	0.56,1.36	0.95	0.82,1.09	0.91	0.74,1.12
2–5	0.94	0.84,1.06	0.82	0.69,0.98	0.99	0.82,1.20	0.94	0.85.1.05	0.84	0.73,0.97
>5	0.99	0.87,1.13	0.89	0.73,1.09	1.04	0.95,1.13	0.99	0.89,1.11	0.90	0.77,1.05
Trend test^b	0.77		0.73		0.31		0.48		0.72
HR in OS/HR in trial^d							1.04	0.91,1.18	1.15	0.97,1.35
Overall HR^d	0.96	0.89, 1.04	0.88	0.78, 0.98	1.03	0.95, 1.11

^aWomen using personal calcium or vitamin D supplements at baseline in the CaD trial are excluded

^bSignificance level (P value) for test of no HR trend across years from CaD initiation categories, coded as 0, 1, 2, respectively

^cOverall HR in the OS divided by that in the CaD trial. This ratio is used as a residual confounding bias correction factor in the OS, in combined trial and cohort study analyses

^dOverall HR is the hazard ratio estimate when the HR is assumed not to depend on years from CaD initiation